Skip to content
DraconiticResearch
Search Sources API
203,948 studies31 sourcesUpdated 19 min ago
Peer-reviewedCC BY-NC-ND 4.0Scientific Data

A Curated Dataset for Question Answering on Laboratory Test Reference Ranges and Interpretation

  • Balu Bhasuran1
  • Qiao Jin2
  • Angelique Deville1
  • Yonghui Wu3
  • Karim Hanna4
  • Zhiyong Lu2
  • Zhe He1,5
5 affiliations
  1. 1School of Information, Florida State University, Tallahassee, Florida USA
  2. 2National Library of Medicine, Bethesda, Maryland USA
  3. 3Department of Health Outcomes and Biomedical Informatics, University of Florida, Gainesville, Florida USA
  4. 4Department of Family Medicine, Morsani College of Medicine, University of South Florida, Tampa, Florida USA
  5. 5Institute for Successful Longevity, Florida State University, Tallahassee, Florida USA
Read full textPDFDOICite

Abstract

Laboratory tests are crucial for diagnosing and managing health conditions, offering essential reference ranges for result interpretation. The diversity of lab tests, influenced by variables such as the specimen type (e.g., blood, urine), gender, age, and other influencing factors like pregnancy, makes automated interpretation challenging. Automated clinical decision support systems attempting to interpret these values must account for such nuances to avoid misdiagnoses or incorrect clinical decisions. In this regard, we present LabQAR (Laboratory Question Answering with Reference Ranges), a manually curated dataset comprising 550 lab test reference ranges derived from authoritative medical sources, encompassing 363 unique lab tests and including multiple-choice questions with annotations on reference ranges, specimen types, and other factors impacting interpretation. We also assess the performance of several large language models (LLMs), including LLaMA 3.1, GatorTronGPT, GPT-3.5 turbo, GPT-4 turbo, and GPT-4o, in predicting reference ranges and classifying results as normal, low, or high. The findings indicate that GPT-4o outperforms other models, showcasing the potential of LLMs in clinical decision support.

Related

  1. A mini-review of medical question answering for Bangla: the methodological landscape and cross-lingual perspectives.
    Jannatul Somiya Mahmud, Khadija Akter Ratri, Md Hamid Hosen +5
    Frontiers in digital healthPeer-reviewedCC BYFull textEpidemiology & public healthEurope PMCPMC
  2. Evidence-based clinical question answering using retrieval-augmented generation (RAG) and open-source large language models
    Ayesha Qasim, Jesus Jaime Moreno Escobar, Christian E. Maldonado-Sifuentes +3
    Scientific ReportsPeer-reviewedCC BY 4.0Full textEpidemiology & public healthCrossref
  3. A generalizable pipeline for specialty-curated postmarket device surveillance: a reference implementation in neurosurgery
    Kashif Qureshi, Egemen Gok, Mohsin Fayaz +1
    Scientific ReportsPeer-reviewedCC BY-NC-ND 4.0Full textPharmacology & toxicologyCrossref
  4. Do question type and topic affect the performance of large language models in answering periodontology questions? A comparative study
    Hanife Merva Parlak
    BMC Medical EducationPeer-reviewedCC BYFull textEpidemiology & public healthDOAJ
  5. OncoAgentic-RAG: An evidence-grounded multi-agent hybrid retrieval-augmented system for patient-centric clinical oncology question answering
    S Jayasrilakshmi, Ansuman Mahapatra, Ankit Kujur +3
    Digital HealthPeer-reviewedCC BYFull textEpidemiology & public healthDOAJ

Cite

BibTeXRIS
BibTeX
@article{bhasuran2026curated,
  title = {{A Curated Dataset for Question Answering on Laboratory Test Reference Ranges and Interpretation}},
  author = {Balu Bhasuran and Qiao Jin and Angelique Deville and Yonghui Wu and Karim Hanna and Zhiyong Lu and Zhe He},
  journal = {Scientific Data},
  year = {2026},
  month = jun,
  doi = {10.1038/s41597-026-07554-z},
  url = {https://doi.org/10.1038/s41597-026-07554-z},
  note = {CC BY-NC-ND 4.0},
}
RIS
TY  - JOUR
TI  - A Curated Dataset for Question Answering on Laboratory Test Reference Ranges and Interpretation
AU  - Balu Bhasuran
AU  - Qiao Jin
AU  - Angelique Deville
AU  - Yonghui Wu
AU  - Karim Hanna
AU  - Zhiyong Lu
AU  - Zhe He
PY  - 2026
DA  - 2026/06/10
JO  - Scientific Data
DO  - 10.1038/s41597-026-07554-z
UR  - https://doi.org/10.1038/s41597-026-07554-z
AB  - Laboratory tests are crucial for diagnosing and managing health conditions, offering essential reference ranges for result interpretation. The diversity of lab tests, influenced by variables such as the specimen type (e.g., blood, urine), gender, age, and other influencing factors like pregnancy, makes automated interpretation challenging. Automated clinical decision support systems attempting to interpret these values must account for such nuances to avoid misdiagnoses or incorrect clinical decisions. In this regard, we present LabQAR (Laboratory Question Answering with Reference Ranges), a manually curated dataset comprising 550 lab test reference ranges derived from authoritative medical sources, encompassing 363 unique lab tests and including multiple-choice questions with annotations on reference ranges, specimen types, and other factors impacting interpretation. We also assess the performance of several large language models (LLMs), including LLaMA 3.1, GatorTronGPT, GPT-3.5 turbo, GPT-4 turbo, and GPT-4o, in predicting reference ranges and classifying results as normal, low, or high. The findings indicate that GPT-4o outperforms other models, showcasing the potential of LLMs in clinical decision support.
SN  - 2052-4463
ER  -